Abstract

BACKGROUND: Both sleep-duration and C-reactive protein (CRP) are useful predictors of coronary heart disease (CHD). The increased CRP level is associated with the unusual sleep-duration. However, it is unclear whether CRP impacts the CHD risk prediction of sleep-duration. METHODS AND RESULTS: A total of 3381 individuals from the Framingham Offspring Study, aged 30+, CHD-free, and without missing measurement of CRP and sleep-duration and being followed to the end of 2007 were included in this analysis. Hazards ratio (HR) from the Cox regression models was used to evaluate the strength of association between the sleep-duration, CRP levels, and risk of incident CHD. Compared to sleep-duration 7-8 h (n = 2512) after adjusting for age and gender, the HR (95% CL) of incident CHD were 1.42 (1.15, 1.76, p < 0.005) for sleep-duration ≤6 h (n = 588) and 1.23 (0.90, 1.70, p < 0.2) for sleep-duration ≥9 h (n = 281), respectively. A further adjustment for other covariates including CRP did not change the CHD risk association. When subjects were categorized into 9 groups by sleep-duration (≤6, 7-8, and ≥9 h) and CRP levels (<1, 1-3, and ≥3 mg/L), and compared to those whose sleep-duration was 7-8 h and CRP levels were <1 mg/L, the HRs of CHD were similar for sleep-duration in ≤6 h or ≥9 h categories regardless of their CRP levels. The increased CRP levels, however, did show an increased risk for CHD when sleep-duration was 7-8 h CONCLUSION: The levels of CRP do not significantly attenuate the association between sleep duration and incident CHD.